Scholarlink

Specialized acoustic imaging technology is used for Seafloor Mapping by reflecting sound from the seafloor to create high-resolution grayscale images using Side-Scan Sonar (SSS). This technique operates effectively in low visibility conditions, minimizes disturbances to marine life, and preserves sensitive underwater environments, making it essential for underwater surveys. However, merging sonar images from large marine areas into continuous maps remains a challenge due to inconsistencies in vessel position, heading, and sensor speed. This research presents a method for seamlessly integrating SSS images from XTF files using Geodesic Distance calculations, Heading Thresholding, and Sensor Speed Continuity Checks to identify optimal file pairs. Crossfade blending minimizes artifacts and intensity inconsistencies, while metadata processing and header matching ensure consistency before merging. Experimental results show a 53.33% file matching accuracy, which is still not good enough. But an efficient processing time of 1.28 s demonstrates suitability for large-scale sonar data integration. Data continuity is confirmed through the mean difference (246.82 - 248.60) and standard deviation (250.27 - 251.09), ensuring smooth transitions. Preprocessing significantly improves data stability, reducing variability (SD: ~ 395 to ~ 0.65–0.67) and maintaining minimal error (MAE: 0.2697, RMSE: 0.3851, Percentage Error: 1.68%). This method enhances underwater mapping by ensuring high-quality sonar image fusion, reducing discrepancies, and improving integration efficiency. Future improvements will focus on adaptive crossfade blending and refining file matching algorithms to enhance image quality and alignment in large-scale seafloor mapping applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhanced Geodesic Distance-Based Fusion of Side-Scan Sonar Images for Seamless Seafloor Mapping

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Nattamon Phalajivin,
  • Nattapat Patchsuwan,
  • Parkpoom Chaisiriprasert

摘要

Specialized acoustic imaging technology is used for Seafloor Mapping by reflecting sound from the seafloor to create high-resolution grayscale images using Side-Scan Sonar (SSS). This technique operates effectively in low visibility conditions, minimizes disturbances to marine life, and preserves sensitive underwater environments, making it essential for underwater surveys. However, merging sonar images from large marine areas into continuous maps remains a challenge due to inconsistencies in vessel position, heading, and sensor speed. This research presents a method for seamlessly integrating SSS images from XTF files using Geodesic Distance calculations, Heading Thresholding, and Sensor Speed Continuity Checks to identify optimal file pairs. Crossfade blending minimizes artifacts and intensity inconsistencies, while metadata processing and header matching ensure consistency before merging. Experimental results show a 53.33% file matching accuracy, which is still not good enough. But an efficient processing time of 1.28 s demonstrates suitability for large-scale sonar data integration. Data continuity is confirmed through the mean difference (246.82 - 248.60) and standard deviation (250.27 - 251.09), ensuring smooth transitions. Preprocessing significantly improves data stability, reducing variability (SD: ~ 395 to ~ 0.65–0.67) and maintaining minimal error (MAE: 0.2697, RMSE: 0.3851, Percentage Error: 1.68%). This method enhances underwater mapping by ensuring high-quality sonar image fusion, reducing discrepancies, and improving integration efficiency. Future improvements will focus on adaptive crossfade blending and refining file matching algorithms to enhance image quality and alignment in large-scale seafloor mapping applications.